Parallelizing Ant Colony Optimization for travelling

salesman problem Over Hadoop Map-Reduce

The travelling salesman problem (TSP) is one of the most intensively studied problems in
optimization. Loosely speaking, given a set of cities on a map, the problem consists of finding a tour
that goes through each city exactly once and ends in the same city it started with. There has been
much research done on finding efficient heuristics to get provably optimal and close to optimal
solutions to TSP problems. Recently a polynomial time approximation scheme was discovered for
the Euclidian TSP problem. With parallel and cloud computing gaining prominence in the past few
years, finding efficient parallelization of existing algorithms will gain importance. Furthermore, given
the push towards cloud computing, it will become increasingly necessary to adopt algorithms to
existing cluster computing frameworks like MapReduce. This project deals with Parallelizing Ant
Colony Optimization for travelling salesman problem Over Hadoop Map-Reduce.Theres already
been much work done in parallelizing TSP heuristics. However there doesn't seem to be any
literature on using advantage of existing cluster computing architectures for solving TSP. In this
project we explore the problem of parallelizing the Ant Colony Optimization algorithm for TSP on the
MapReduce cluster computing framework. We focus on parallelizing Ant Colony Optimization for
travelling salesman problem Over Hadoop Map-Reduce.
Generally, for a TSP solver, one either tries to obtain a provably optimal solution of one tries to get a
solution as close to the optimum as possible without actually proving that the solution is close to the
optimum. While the former goal has an advantage in that it gives a guarantee of the quality of the
solution, it is generally very slow and infeasible to apply to instances of a large size. Thus we opt for
the second goal. We use an algorithm termed the Ant Colony Optimization algorithm that simulates
the way ants find the shortest route to a food source. There already exist sequential versions of this
algorithm. We aim to parallelize this algorithm over Hadoop MapReduce and thereby improve its
performance.


In computer science and operations research, the ant colony optimization (ACO) algorithm is a
probabilistic technique for solving computational problems which can be reduced to finding good
paths through graphs. Ant colony optimization algorithms have been applied to many combinatorial
optimization problems, ranging from quadratic assignment to protein folding or routing vehicles and a
lot of derived methods have been adapted to dynamic problems in real variables, stochastic
problems, multi-targets and parallel implementations. It has also been used to produce near-optimal
solutions to the travelling salesman problem.


They have an advantage over simulated annealing and genetic algorithm approaches of similar
problems when the graph may change dynamically; the ant colony algorithm can be run continuously
and adapt to changes in real time. This is of interest in network routing and urban transportation
systems.

As a very good example, ant colony optimization algorithms have been used to produce near-
optimal solutions to the travelling salesman problem. The first ACO algorithm was called the Ant
system and it was aimed to solve the travelling salesman problem, in which the goal is to find the
shortest round-trip to link a series of cities. The general algorithm is relatively simple and based on a
set of ants, each making one of the possible round-trips along the cities. At each stage, the ant
chooses to move from one city to another according to some rules: